CN110148230A - A kind of vehicle load-carrying prediction technique based on LSTM neural network - Google Patents
A kind of vehicle load-carrying prediction technique based on LSTM neural network Download PDFInfo
- Publication number
- CN110148230A CN110148230A CN201910416481.7A CN201910416481A CN110148230A CN 110148230 A CN110148230 A CN 110148230A CN 201910416481 A CN201910416481 A CN 201910416481A CN 110148230 A CN110148230 A CN 110148230A
- Authority
- CN
- China
- Prior art keywords
- carrying
- data
- load
- vehicle
- neural network
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
- G06Q10/083—Shipping
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C5/00—Registering or indicating the working of vehicles
- G07C5/08—Registering or indicating performance data other than driving, working, idle, or waiting time, with or without registering driving, working, idle or waiting time
- G07C5/0808—Diagnosing performance data
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0108—Measuring and analyzing of parameters relative to traffic conditions based on the source of data
- G08G1/0112—Measuring and analyzing of parameters relative to traffic conditions based on the source of data from the vehicle, e.g. floating car data [FCD]
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Economics (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Human Resources & Organizations (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Entrepreneurship & Innovation (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Game Theory and Decision Science (AREA)
- Control Of Transmission Device (AREA)
Abstract
The vehicle load-carrying prediction technique based on LSTM neural network that the present invention provides a kind of, acquires vehicle data by car-mounted terminal;Vehicle data is integrated, data cleansing and standardization, obtains training data;It is trained by shot and long term Memory Neural Networks model, obtains load-carrying model;It is predicted the load-carrying model after prediction data input training to obtain load-carrying data.A kind of vehicle load-carrying prediction technique based on LSTM neural network provided by the invention utilizes big data computing platform, application based on neural network model in prediction field, in conjunction with longitudinal dynamics vehicle mass accounting equation, it can be realized and carload in real time, accurately predict.
Description
Technical field
The vehicle load-carrying prediction technique based on LSTM neural network that the present invention relates to a kind of, belongs to car networking and logistics transportation
Technical field.
Background technique
There are mainly two types of for the method for the prediction of tradition load-carrying at present:
Method one: complete vehicle quality is estimated using forgetting factor recurrent least square method based on longitudinal dynamics.
Disadvantage: not considering gear shift operation, carried out in shift process estimation will lead to algorithm estimated accuracy it is not high and convergence speed
Degree is slow.And due to brake control power, air drag in reality, the factors real-time change such as force of rolling friction is difficult accurately to obtain,
Also the precision of prediction is affected to a certain extent.Its prediction result and true value comparison error range be difficult control 15% with
It is interior.
Method two: using sensor is installed on automobile, vehicle live load is estimated by sensor deformation.
Disadvantage: it requires that sensor is additionally installed on every trolley, the price of a common vehicle load sensor just exists
It is 10000 yuans or more, with high costs, and the sensor based on deformation is easy aging, will increase later maintenance cost, it is right
The calculating of load-carrying precision can also have an impact.
Summary of the invention
In order to solve the deficiencies in the prior art, increases predictablity rate, reduce cost, the present invention provides one kind to be based on
The vehicle load-carrying prediction technique of LSTM neural network, using big data computing platform, based on neural network model in prediction field
Application, can in the case where additionally not adding vehicle load sensor in conjunction with longitudinal dynamics vehicle mass accounting equation
It realizes and carload in real time, accurately predict, prediction result error is within 5%.
The present invention is that technical solution used by solving its technical problem is: being provided a kind of based on LSTM neural network
Vehicle load-carrying prediction technique, comprising the following steps:
(1) vehicle data is acquired by car-mounted terminal;
(2) vehicle data is integrated, data cleansing and standardization, obtains training data;
(3) training data is trained by shot and long term Memory Neural Networks model, obtains load-carrying model;
(4) the load-carrying model after prediction data input training is predicted, obtains the load-carrying prediction result of target vehicle
Collection, takes the median of result set as final prediction result.
Step (1) described vehicle data include the moment, speed, engine speed, clutch switch, gas pedal aperture and
Brake switch and torque or torque percentage.
Step (2) integration is that every vehicle data at continuous two moment is integrated into one group of characteristic index number respectively
According to.
Step (2) described data cleansing is to be filtered according to the following conditions to vehicle data, for the vehicle at each moment
Data, meet following all conditions and just retain:
Condition one: the difference of the speed of later moment in time and previous moment is greater than 0;
Condition two: the effective range of engine speed n is 700 revs/min of 2500 revs/min of < n <;
Condition three: the effective range of torque T e is 1000Nm < Te < 2200Nm;
Condition four: brake switch fbswitch=0;
Condition five: time interval deltaT≤5s of later moment in time and previous moment;
Condition six: Isolating Switch lhswitch=0.
Step (2) is standardized using z-score standardized method.
Step (3) is during shot and long term Memory Neural Networks model is trained, using RMSE as evaluation criterion,
Wherein training set error loss train_loss, verifying collection error loss val_loss and prediction and error loss pre_loss are got over
It is small, indicate that the fitting effect of shot and long term Memory Neural Networks model is better.
For load-carrying model, the Outside Access interface of load-carrying model is set, is called for third party and is predicted to obtain load-carrying
Data.
For load-carrying model, it is transplanted to car-mounted terminal, for collected vehicle data directly to be inputted load-carrying mould
Type, the load-carrying data that output prediction obtains.
The present invention is based on beneficial effects possessed by its technical solution to be:
(1) a kind of vehicle load-carrying prediction technique calculating process based on LSTM neural network provided by the invention does not depend on brake
Vehicle brake force, air drag, force of rolling friction etc. be difficult to acquire with environmental factor delta data, by experiment, prediction result
Error is within 5%, and compared with traditional least square method error range is more than 15%, precision of prediction is obviously improved;
(2) a kind of vehicle load-carrying prediction technique based on LSTM neural network provided by the invention only acquires car-mounted terminal hair
The data sent save manufacturing cost and later maintenance cost without additionally installing the relevant sensor of any quality;
(3) a kind of vehicle load-carrying prediction technique based on LSTM neural network provided by the invention can be to vehicle in terms of transport
Carry out Payload Monitoring And Control during transportation, rationally utilize (scheduling) load resource, monitor cargo whether reach designated place with
And early warning is carried out to overload, overload;
(4) can there is aspect after a kind of vehicle load-carrying prediction technique based on LSTM neural network provided by the invention is on sale
Effect, which understands the failure defining problem that vehicle is under warranty and overload, influences vehicle component service life bring;Whole
Vehicle research and development aspect, it is whole to improve by performance and fuel-economy sex expression of the big data mining analysis vehicle under different loads
Vehicle level provides data foundation;
(5) it is provided by the invention it is a kind of based on the vehicle load-carrying prediction technique of LSTM neural network in vehicle performance optimization side
Face can pass through the prediction of real-time load and interact result and entire car controller, and it is excellent to reach controller control strategy dynamic
Change, improves vehicle economical operation benefit and safety.
Detailed description of the invention
Fig. 1 is the schematic diagram data after integration.
Specific embodiment
Present invention will be further explained below with reference to the attached drawings and examples.
The vehicle load-carrying prediction technique based on LSTM neural network that the present invention provides a kind of, comprising the following steps:
(1) vehicle data is acquired by car-mounted terminal (T-BOX), including moment, speed, engine speed, clutch open
Pass, gas pedal aperture and brake switch and torque or torque percentage;
(2) vehicle data is integrated, data cleansing and standardization, obtains training data;
This step is mainly prepared work to data.By longitudinal dynamics formula:
Wherein M indicates the total quality of automobile, and v indicates automobile driving speed (can acquire data), accelerationTeTable
Show torque (can acquire data) of the engine action on flywheel,Indicate total gear ratio, wherein rwIndicate wheel half
Diameter (vehicle quiescent value), gdFor gear ratio (vehicle quiescent value), gfIndicate driving than (vehicle quiescent value), JeIndicate rotary inertia
(vehicle quiescent value), ω indicate engine speed (can acquire data), FfbIndicate that the braking friction on wheel (can acquire number
According to),Wherein CdIt is coefficient of air resistance, ρ indicates atmospheric density, and A is automotive windshield area, Fgrade=
Mg (μ cos β+sin β) is the resultant force generated by the gradient and rolling friction, and wherein μ indicates that road surface rolling friction force coefficient, β indicate
The gradient, β>0 are to go up a slope, and β<0 is descending, and β=0 is no gradient.
By longitudinal dynamics formula it is found that terminal acquisition data in, information relevant to quality has: the time, speed,
Transmitter revolving speed, torque, brake switch, clutch switch and gas pedal aperture.Since the calculating of acceleration at least needs two
The reported data at a moment, therefore data are integrated first, every vehicle data at continuous two moment is integrated into respectively
One group of characteristic index data, i.e., every data line shown in FIG. 1.
Then data cleansing is carried out, some wrong data are filtered out and generates the dirty data of interference to calculated result.According to
The following conditions are filtered vehicle data, for the vehicle data at each moment, meet following all conditions and just retain:
Condition one: the difference of the speed of later moment in time and previous moment is greater than 0;
Condition two: the effective range of engine speed n is 700 revs/min of 2500 revs/min of < n <;
Condition three: the effective range of torque T e is 1000Nm < Te < 2200Nm;
Condition four: brake switch fbswitch=0;
Condition five: time interval deltaT≤5s of later moment in time and previous moment;
Condition six: Isolating Switch lhswitch=0.
The data of acquisition include multiple indexs, since the property of each index is different, have different dimension and the order of magnitude.It examines
The level difference considered between certain indexs is larger, if directly analyzed with original index value, it is higher will to protrude numerical value
Effect of the index in comprehensive analysis, the opposite effect for weakening the horizontal lower index of numerical value.Therefore, in order to guarantee the reliable of result
Property, it needs to be standardized original index data.Standard is carried out using data of the z-score standardized method to acquisition
Change processing, mean value (mean) and standard deviation (standard deviation) of this method based on initial data carry out data
Standardization, i.e. new data=(former data-mean value)/standard deviation.
(3) shot and long term Memory Neural Networks model is established, input layer number is 8, and output layer neuron number is
1, activation primitive ReLU, mathematic(al) representation are as follows: f (x)=max (α x, x), wherein α=0.01, x are input vector, f (x)
For output vector, activation primitive is a non-linear transfer function of neural computing pilot process, for realizing one-to-one
Transformation, i.e., mapped with each component of the identical function to input vector, obtain output vector.Optimizer uses
Adam.Training data is trained by shot and long term Memory Neural Networks model, obtains load-carrying model.To make prediction result more
It is accurate to add, and reduces influence of the item forecast result to whole prediction result, when being predicted, prediction data takes continuous time period
Interior data set (prediction data is no less than 50), is entered into load-carrying model and obtains prediction result collection, take in result set
Digit is as final prediction result and output.
In training process, evaluation criterion is used as using RMSE (Root Mean Squared Error, root-mean-square error),
Parameter optimization is also to carry out on the basis of RMSE, and the loss function etc. of load-carrying prediction is also all on the basis of RMSE
It is iterated optimization, error costing bio disturbance formula are as follows:
Wherein tiIndicate target value (preset value) corresponding to i-th group of characteristic index data of input, yiIndicate i-th group of input
Predicted value corresponding to characteristic index data (result of output) obtains corresponding error by the difference of input data set and damages
It loses.Training set error is obtained when input data set is training set loses train_loss;When input data is verifying collection, obtain
Val_loss is lost to verifying collection error;When input data set is forecast set, forecast set error loss pre_loss is obtained.Respectively
The value of loss is lower, and the effect for illustrating models fitting is better.
(4) in view of prediction result concentration has that individual data is serious bigger than normal or less than normal, in order to preferably show
Load-carrying model after prediction data input training is predicted, obtains target vehicle by the mean level of this group of prediction result
Load-carrying prediction result collection exports final prediction result using the method for taking median to result set, and the computation rule of median is such as
Under:
The data concentrated for prediction result: X1..., XN, arranged by sequence from small to large are as follows: X(1)...,
X(N).When N is odd number, load value m is predicted0.5=X(N+1)/2, when N is even number, predict load value
For load-carrying model, load-carrying model is arranged is externally based on ICP/IP protocol stack addressing interface, calls for third party
It is predicted to obtain load-carrying data.By interface, predictive data set is uploaded, selects prediction data characteristic series, characteristic series must wrap
Have containing information: calling time in data, speed, engine speed, clutch switch, gas pedal aperture, brake switch, torque or
Person's torque percentage.Then it calls and has saved prediction model, finally obtain prediction result.
For load-carrying model, it can also be transplanted to car-mounted terminal (T-BOX), by T-BOX, acquire predictive data set,
Prediction data characteristic series are selected, characteristic series must have comprising information: call time in data, speed, engine speed, clutch are opened
It closes, gas pedal aperture, brake switch, collected vehicle data is directly inputted load-carrying mould by torque or torque percentage
Type, the load-carrying data finally predicted, and the interaction of direct and vehicle control device, realize optimizing entire vehicle controller strategy.
A kind of vehicle load-carrying prediction technique based on LSTM neural network provided by the invention, by neural network model with
Longitudinal dynamics combine, and longitudinal dynamics provide theories integration for selection, the integration of neural network model feature column data,
LSTM network provides the machine learning algorithm with memory function, and the data reported using car-mounted terminal no longer need to pass by other
Sensor is enough quick and precisely to identify vehicle load.
Claims (8)
1. a kind of vehicle load-carrying prediction technique based on LSTM neural network, it is characterised in that the following steps are included:
(1) vehicle data is acquired by car-mounted terminal;
(2) vehicle data is integrated, data cleansing and standardization, obtains training data;
(3) training data is trained by shot and long term Memory Neural Networks model, obtains load-carrying model;
(4) the load-carrying model after prediction data input training is predicted, obtains the load-carrying prediction result collection of target vehicle, takes
The median of result set is as final prediction result.
2. the vehicle load-carrying prediction technique according to claim 1 based on LSTM neural network, it is characterised in that: step
(1) vehicle data includes moment, speed, engine speed, clutch switch, gas pedal aperture and brake switch, with
And torque or torque percentage.
3. the vehicle load-carrying prediction technique according to claim 1 based on LSTM neural network, it is characterised in that: step
(2) integration is that every vehicle data at continuous two moment is integrated into one group of characteristic index data respectively.
4. the vehicle load-carrying prediction technique according to claim 1 based on LSTM neural network, it is characterised in that: step
(2) data cleansing is to be filtered according to the following conditions to vehicle data, for the vehicle data at each moment, meet with
Lower all conditions just retain:
Condition one: the difference of the speed of later moment in time and previous moment is greater than 0;
Condition two: the effective range of engine speed n is 700 revs/min of 2500 revs/min of < n <;
Condition three: the effective range of torque T e is 1000Nm < Te < 2200Nm;
Condition four: brake switch fbswitch=0;
Condition five: time interval deltaT≤5s of later moment in time and previous moment;
Condition six: Isolating Switch lhswitch=0.
5. the vehicle load-carrying prediction technique according to claim 1 based on LSTM neural network, it is characterised in that: step
(2) it is standardized using z-score standardized method.
6. the vehicle load-carrying prediction technique according to claim 1 based on LSTM neural network, it is characterised in that: step
(3) during shot and long term Memory Neural Networks model is trained, using RMSE as evaluation criterion, wherein training set is missed
Differential loss loses train_loss, verifying collection error loss val_loss and prediction and error loss pre_loss is smaller, indicates length
The phase fitting effect of Memory Neural Networks model is better.
7. the vehicle load-carrying prediction technique according to claim 1 based on LSTM neural network, it is characterised in that: for carrying
The Outside Access interface of load-carrying model is arranged in molality type, calls for third party and is predicted to obtain load-carrying data.
8. the vehicle load-carrying prediction technique according to claim 1 based on LSTM neural network, it is characterised in that: for carrying
Molality type, is transplanted to car-mounted terminal, and for collected vehicle data directly to be inputted load-carrying model, output prediction is obtained
Load-carrying data.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910416481.7A CN110148230B (en) | 2019-05-20 | 2019-05-20 | Whole vehicle load prediction method based on LSTM neural network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910416481.7A CN110148230B (en) | 2019-05-20 | 2019-05-20 | Whole vehicle load prediction method based on LSTM neural network |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110148230A true CN110148230A (en) | 2019-08-20 |
CN110148230B CN110148230B (en) | 2021-05-25 |
Family
ID=67591844
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910416481.7A Active CN110148230B (en) | 2019-05-20 | 2019-05-20 | Whole vehicle load prediction method based on LSTM neural network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110148230B (en) |
Cited By (19)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111027146A (en) * | 2019-12-30 | 2020-04-17 | 行蜂科技(深圳)有限责任公司 | Vehicle load dynamic real-time calculation method |
CN111090959A (en) * | 2019-12-31 | 2020-05-01 | 湖南行必达网联科技有限公司 | Vehicle load spectrum acquisition method and system |
CN111157092A (en) * | 2020-01-02 | 2020-05-15 | 深圳市汉德网络科技有限公司 | Vehicle-mounted weighing automatic calibration method and computer readable storage medium |
CN111209695A (en) * | 2019-12-30 | 2020-05-29 | 浙江大学 | LSTM-based structural dynamic response prediction method |
CN111483277A (en) * | 2020-04-26 | 2020-08-04 | 武汉大学 | Tire underpressure identification method and system based on machine learning and storage medium |
CN112373484A (en) * | 2020-11-26 | 2021-02-19 | 同济大学 | Method for acquiring vehicle mass dynamics based on feedforward neural network |
CN112441016A (en) * | 2019-09-05 | 2021-03-05 | 百度(美国)有限责任公司 | Gear-based vehicle load inference system |
CN112446402A (en) * | 2019-09-03 | 2021-03-05 | 顺丰科技有限公司 | Loading rate identification method and device, computer equipment and storage medium |
CN112766771A (en) * | 2021-01-26 | 2021-05-07 | 中国重汽集团济南动力有限公司 | SVM (support vector machine) -based whole heavy-duty automobile mass prediction method |
CN112819031A (en) * | 2021-01-04 | 2021-05-18 | 中国汽车技术研究中心有限公司 | Vehicle-mounted weight prediction method and system, electronic device and medium |
CN112949187A (en) * | 2021-03-05 | 2021-06-11 | 株洲齿轮有限责任公司 | Vehicle mass calculation method |
CN113849910A (en) * | 2021-09-23 | 2021-12-28 | 四川大学 | Dropout-based BiLSTM network wing resistance coefficient prediction method |
CN113989534A (en) * | 2021-10-15 | 2022-01-28 | 厦门大学 | Method and system for identifying load between bus stations based on multi-source operation data |
CN114036830A (en) * | 2021-11-03 | 2022-02-11 | 厦门大学 | Bus load prediction method and device based on multi-view ensemble learning |
CN114386892A (en) * | 2021-11-30 | 2022-04-22 | 江铃汽车股份有限公司 | Transportation vehicle load management method, system, storage medium and equipment |
CN114435378A (en) * | 2020-11-03 | 2022-05-06 | 陕西汽车集团有限责任公司 | Pure electric vehicle whole vehicle mass estimation method based on neural network |
CN114912073A (en) * | 2021-02-07 | 2022-08-16 | 北京潼荔科技有限公司 | Method for measuring and calculating single-tire load, whole-vehicle load and gravity center position of vehicle |
CN117315943A (en) * | 2023-11-28 | 2023-12-29 | 湖南省交通科学研究院有限公司 | Monitoring analysis and early warning method and system for overrun transportation violations |
CN117494585A (en) * | 2023-12-29 | 2024-02-02 | 中汽研汽车检验中心(天津)有限公司 | Commercial vehicle actual load prediction method based on mutual information and data blurring |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103264669A (en) * | 2013-05-31 | 2013-08-28 | 吉林大学 | Heavy vehicle weight real-time identification method based on CAN information and function principle |
CN106844949A (en) * | 2017-01-18 | 2017-06-13 | 清华大学 | A kind of training method for realizing the controllable two-way LSTM models of locomotive section |
WO2017223009A1 (en) * | 2016-06-23 | 2017-12-28 | Microsoft Technology Licensing, Llc | Multi-domain joint semantic frame parsing |
CN109255094A (en) * | 2018-08-10 | 2019-01-22 | 重庆邮电大学 | Commercial truck quality estimation method based on SVR_Adaboost innovatory algorithm |
CN109635830A (en) * | 2018-10-24 | 2019-04-16 | 吉林大学 | For estimating the screening technique of the valid data of car mass |
-
2019
- 2019-05-20 CN CN201910416481.7A patent/CN110148230B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103264669A (en) * | 2013-05-31 | 2013-08-28 | 吉林大学 | Heavy vehicle weight real-time identification method based on CAN information and function principle |
WO2017223009A1 (en) * | 2016-06-23 | 2017-12-28 | Microsoft Technology Licensing, Llc | Multi-domain joint semantic frame parsing |
CN106844949A (en) * | 2017-01-18 | 2017-06-13 | 清华大学 | A kind of training method for realizing the controllable two-way LSTM models of locomotive section |
CN109255094A (en) * | 2018-08-10 | 2019-01-22 | 重庆邮电大学 | Commercial truck quality estimation method based on SVR_Adaboost innovatory algorithm |
CN109635830A (en) * | 2018-10-24 | 2019-04-16 | 吉林大学 | For estimating the screening technique of the valid data of car mass |
Cited By (32)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112446402A (en) * | 2019-09-03 | 2021-03-05 | 顺丰科技有限公司 | Loading rate identification method and device, computer equipment and storage medium |
CN112446402B (en) * | 2019-09-03 | 2024-03-26 | 顺丰科技有限公司 | Loading rate identification method, loading rate identification device, computer equipment and storage medium |
CN112441016B (en) * | 2019-09-05 | 2024-08-13 | 百度(美国)有限责任公司 | Gear-based vehicle load inference system |
CN112441016A (en) * | 2019-09-05 | 2021-03-05 | 百度(美国)有限责任公司 | Gear-based vehicle load inference system |
CN111209695A (en) * | 2019-12-30 | 2020-05-29 | 浙江大学 | LSTM-based structural dynamic response prediction method |
CN111027146B (en) * | 2019-12-30 | 2023-11-24 | 行蜂科技(深圳)有限责任公司 | Dynamic real-time calculation method for vehicle load |
CN111027146A (en) * | 2019-12-30 | 2020-04-17 | 行蜂科技(深圳)有限责任公司 | Vehicle load dynamic real-time calculation method |
CN111090959A (en) * | 2019-12-31 | 2020-05-01 | 湖南行必达网联科技有限公司 | Vehicle load spectrum acquisition method and system |
CN111090959B (en) * | 2019-12-31 | 2023-06-20 | 湖南行必达网联科技有限公司 | Vehicle load spectrum acquisition method and system |
CN111157092A (en) * | 2020-01-02 | 2020-05-15 | 深圳市汉德网络科技有限公司 | Vehicle-mounted weighing automatic calibration method and computer readable storage medium |
CN111483277A (en) * | 2020-04-26 | 2020-08-04 | 武汉大学 | Tire underpressure identification method and system based on machine learning and storage medium |
CN114435378B (en) * | 2020-11-03 | 2024-08-09 | 陕西汽车集团股份有限公司 | Whole-vehicle quality estimation method for pure electric vehicle based on neural network |
CN114435378A (en) * | 2020-11-03 | 2022-05-06 | 陕西汽车集团有限责任公司 | Pure electric vehicle whole vehicle mass estimation method based on neural network |
CN112373484B (en) * | 2020-11-26 | 2021-12-31 | 同济大学 | Method for acquiring vehicle mass dynamics based on feedforward neural network |
CN112373484A (en) * | 2020-11-26 | 2021-02-19 | 同济大学 | Method for acquiring vehicle mass dynamics based on feedforward neural network |
CN112819031A (en) * | 2021-01-04 | 2021-05-18 | 中国汽车技术研究中心有限公司 | Vehicle-mounted weight prediction method and system, electronic device and medium |
CN112819031B (en) * | 2021-01-04 | 2022-06-17 | 中国汽车技术研究中心有限公司 | Vehicle-mounted weight prediction method and system, electronic device and medium |
CN112766771A (en) * | 2021-01-26 | 2021-05-07 | 中国重汽集团济南动力有限公司 | SVM (support vector machine) -based whole heavy-duty automobile mass prediction method |
CN114912073A (en) * | 2021-02-07 | 2022-08-16 | 北京潼荔科技有限公司 | Method for measuring and calculating single-tire load, whole-vehicle load and gravity center position of vehicle |
CN114912073B (en) * | 2021-02-07 | 2023-11-21 | 北京潼荔科技有限公司 | Method for measuring and calculating single-tire load, whole-vehicle load and gravity center position of vehicle |
CN112949187B (en) * | 2021-03-05 | 2022-06-14 | 株洲齿轮有限责任公司 | Vehicle mass calculation method |
CN112949187A (en) * | 2021-03-05 | 2021-06-11 | 株洲齿轮有限责任公司 | Vehicle mass calculation method |
CN113849910A (en) * | 2021-09-23 | 2021-12-28 | 四川大学 | Dropout-based BiLSTM network wing resistance coefficient prediction method |
CN113989534A (en) * | 2021-10-15 | 2022-01-28 | 厦门大学 | Method and system for identifying load between bus stations based on multi-source operation data |
CN113989534B (en) * | 2021-10-15 | 2024-07-02 | 厦门大学 | Bus inter-station load identification method and system based on multi-source operation data |
CN114036830A (en) * | 2021-11-03 | 2022-02-11 | 厦门大学 | Bus load prediction method and device based on multi-view ensemble learning |
CN114036830B (en) * | 2021-11-03 | 2024-06-11 | 厦门大学 | Bus load prediction method and device based on multi-view integrated learning |
CN114386892A (en) * | 2021-11-30 | 2022-04-22 | 江铃汽车股份有限公司 | Transportation vehicle load management method, system, storage medium and equipment |
CN117315943A (en) * | 2023-11-28 | 2023-12-29 | 湖南省交通科学研究院有限公司 | Monitoring analysis and early warning method and system for overrun transportation violations |
CN117315943B (en) * | 2023-11-28 | 2024-02-06 | 湖南省交通科学研究院有限公司 | Monitoring analysis and early warning method and system for overrun transportation violations |
CN117494585A (en) * | 2023-12-29 | 2024-02-02 | 中汽研汽车检验中心(天津)有限公司 | Commercial vehicle actual load prediction method based on mutual information and data blurring |
CN117494585B (en) * | 2023-12-29 | 2024-04-02 | 中汽研汽车检验中心(天津)有限公司 | Commercial vehicle actual load prediction method based on mutual information and data blurring |
Also Published As
Publication number | Publication date |
---|---|
CN110148230B (en) | 2021-05-25 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110148230A (en) | A kind of vehicle load-carrying prediction technique based on LSTM neural network | |
CN110228470B (en) | Fuel saving rate real-time calculation method based on hidden vehicle model prediction | |
CN110610260A (en) | Driving energy consumption prediction system, method, storage medium and equipment | |
CN112614344B (en) | Hybrid traffic system efficiency evaluation method for automatic driving automobile participation | |
CN103544389B (en) | Autocrane method for diagnosing faults based on fault tree and fuzzy neural network | |
CN105389863B (en) | A kind of vehicle condition index calculation method assessed vehicle condition | |
CN108960426A (en) | Road grade Synthesize estimation system based on BP neural network | |
WO2020082910A1 (en) | Method and device for determining pavement rating, storage medium and automobile | |
CN112883648A (en) | Training method and device for automobile fuel consumption prediction model and computer equipment | |
CN106527124B (en) | Electromagnetic type damper control method based on non-linear neural fuzzy controller | |
CN110990950A (en) | Multi-target train energy-saving optimization method based on hybrid operation mode | |
CN111707343B (en) | Method and device for determining weight of vehicle | |
CN112198799A (en) | High-speed train parking control method and system based on deep learning | |
CN111198032A (en) | Real-time estimation method for automobile mass | |
Liu et al. | The hybrid predictive model of elevator system for energy consumption | |
CN115492928A (en) | Economic, dynamic and safety comprehensive optimal gear shifting rule optimization method | |
CN113954846B (en) | Estimation and system for ramp information in vehicle running | |
RU2709357C2 (en) | Compression algorithms for transmitting messages over vehicle bus with preliminary presentation data | |
Lei et al. | Adaptive gearshift strategy based on generalized load recognition for automatic transmission vehicles | |
CN113987927A (en) | Commercial vehicle-mounted quality detection method | |
CN109886439A (en) | Electronic-control vehicle remote diagnosis system and its diagnostic method | |
CN116118705B (en) | Energy management control method for plug-in hybrid power bus in following scene | |
CN107143649A (en) | A kind of congestion industry and mining city and fluid drive gearshift update the system and its method | |
CN207208040U (en) | A kind of adaptive active breaking systems of vehicle | |
CN110281774A (en) | One kind being based on big data, cloud platform Vehicle oil consumption management system and management method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |